Exploring Quality of Constraints for Assessment in Problem Solving Environments
Jaime Galvez Cordero, Eduardo Guzman De Los Riscos, Ricardo Conejo Muñoz
One of the approaches that has demonstrated by far its efficiency as a tutorial strategy in problem solving learning environments is the Constraint-Based Modeling (CBM). In existing works it has been combined with a data-driven technique for automatic assessment, the Item Response Theory (IRT). The result is a well-founded model for assessing students while solving problems. In this paper a novel technique for studying quality of constraints for this type of assessment is presented. It has been tested with two new systems, an independent component for assessment that implements CBM with IRT, which provides assessment to a new problem solving environment developed to assess the students’ skills in decision-making in project investments. The results of testing our approach and the application of these two systems with undergraduate students are also discussed in this paper.
The final publication is available at Springer via https://doi.org/10.1007/978-3-642-30950-2_41.